Self-regulation is a key aspect of healthy development and impairments in self-regulation are associated with many poor health outcomes. This application addresses the NIH's call for greater consistency and integration in research on self-regulation. Building from a core framework, wherein self-regulation is defined as the recruitment of executive processes (e.g. attention control) to alter prepotent responses (e.g. emotion) we develop a set of mathematical models that aim to unify theoretical perspectives, and to capture the dynamic nature of self-regulation using intensive time-series data. Models are developed and tested using intensive time-series data archived in 5 independent studies of early childhood self-regulation and parent-directed regulation of infant state. In each, behaviors and/or physiology indexing prepotent and executive processes were coded on a second-by-second time scale. Knowledge gained from the unified framework is then used to design and collect new data that provides for tests of predictive validity, developmental differences, and generalization across multiple manifestations of both young children's and parents' self-regulation.